Currently, marginal energy, or peak energy, is produced mainly by gas turbines, operating either in simple cycle or combined cycle configurations. As a result of load demand profile, the gas turbine base systems are cycled up during periods of high demand and cycled down, or turned off, during periods of low demand. This cycling is typically driven by the electrical grid operator under a program called “active grid control”, or AGC. Unfortunately, because industrial gas turbines, which represent the majority of the installed power generation base, were designed primarily for base load operation, a severe penalty is associated with the maintenance cost of that particular unit when they are cycled. For example, a gas turbine that is running base load might go through a normal maintenance cycle once every three years, or 24,000 hours of operation, at a cost of between two million dollars and three million dollars ($2,000,000 to $3,000,000). That same cost could be incurred in one year for a gas turbine that is forced to start up and shut down every day due to the severe penalty associated with the maintenance cost of cycling that particular gas turbine. Also, even aero-derivative engines, which are designed for quick starting capability, may still take ten (10) minutes or longer to deliver the required power when called on. This need to cycle the gas turbine fleet is a major issue, and is becoming more problematic with the increased use of intermittent renewable energy sources on the grid.
Currently the gas turbine engines used at power plants can turn down to approximately 50% of their rated capacity. They do this by closing the inlet guide vanes of the compressor, which reduces the air flow to the gas turbine and in turn reduces fuel flow, as a constant fuel air ratio is desired in the combustion process at all engine operating conditions. The goal of maintaining safe compressor operation and gas turbine exhaust emissions typically limit the level of turn down that can be practically achieved.
One way to safely lower the operating limit of the compressor in current gas turbines is by introducing warm air to the inlet of the gas turbine, typically extracted from a mid-stage bleed port on the compressor. Sometimes, this warm air is introduced into the inlet to prevent icing as well. In either case, when this is done, the work that is done to the air by the compressor is sacrificed in the process for the benefit of being able to operate the compressor safely at a lower air flow, yielding the increased turn down capability. Unfortunately, bleeding air from the compressor has a further negative impact on the efficiency of the overall gas turbine system as the work performed on the air that is bled off is lost. In general, for every 1% of air that is bled off the compressor for this turn down improvement, approximately 2% of the total power output of the gas turbine is lost. Additionally, the combustion system also presents a limit to the system.
The combustion system usually limits the amount that the system can be turned down because as less fuel is added, the flame temperature reduces, increasing the amount of carbon monoxide (“CO”) emissions produced. The relationship between flame temperature and CO emissions is exponential with reducing temperature, consequently, as the gas turbine system gets near the turn-down limit, the CO emissions spike up, so it is important to a maintain a healthy margin from this limit. This characteristic limits all gas turbine systems to approximately 50% turn down capability, or, for a 100 MW gas turbine, the minimum power turn-down that can be achieved is about 50%, or 50 MW. As the gas turbine mass flow is turned down, the compressor and turbine efficiency falls off as well, causing an increase in heat rate of the machine. Some operators are faced with this situation every day and as a result, as the load demand falls, gas turbine plants hit its lower operating limit and the gas turbines have to be turned off, which causes the power plant to incur a tremendous maintenance cost penalty.
Another characteristic of a typical gas turbine is that as the ambient temperature increases, the power output goes down proportionately due to the linear effect of the reduced density as the temperature of air increases. Power output can be down by more than 10% from nameplate power rating during hot days, which is typically when peaking gas turbines are called on most frequently to deliver power.
Another characteristic of typical gas turbines is that air that is compressed and heated in the compressor section of the gas turbine is ducted to different portions of the gas turbine's turbine section where it is used to cool various components. This air is typically called turbine cooling and leakage air (hereinafter “TCLA”) a term that is well known in the art with respect to gas turbines. Although heated from the compression process, TCLA air is still significantly cooler than the turbine temperatures, and thus is effective in cooling those components in the turbine downstream of the compressor. Typically 10% to 15% of the air that enters the inlet of the compressor bypasses the combustor and is used for this process. Thus, TCLA is a significant penalty to the performance of the gas turbine system.
Other power augmentation systems, like inlet chilling for example, provide cooler inlet conditions, resulting in increased air flow through the gas turbine compressor, and the gas turbine output increases proportionately. For example, if inlet chilling reduces the inlet conditions on a hot day such that the gas turbine compressor has 5% more air flow, the output of the gas turbine will also increase by 5%. As ambient temperatures drops, inlet chilling becomes less effective, since the air is already cold. Therefore, inlet chilling power increase is maximized on hot days, and tapers off to zero at approximately 45° F. ambient temperature days.
In power augmentation systems such as the one discussed in U.S. Pat. No. 6,305,158 to Nakhamkin (the “'158 patent”), there are three basic modes of operation defined, a normal mode, charging mode, and an air injection mode, but it is limited by the need for an electrical generator that has the capacity to deliver power “exceeding the full rated power” that the gas turbine system can deliver. The fact that this patent has been issued for more than ten (10) years and yet there are no known applications of it at a time of rapidly rising energy costs is proof that it does not address the market requirements. First of all, it is very expensive to replace and upgrade the electrical generator so it can deliver power “exceeding the full rated power” that the gas turbine system can currently deliver. Also, although the injection option as disclosed in the '158 patent provides power augmentation, it takes a significant amount of time to start and get on line to the electrical grid. This makes application of the '158 patent impractical in certain markets like spinning reserve, where the power increase must occur in a matter of seconds, and due to do the need for the large auxiliary compressor in these types of systems, that takes too long to start.
Another drawback is that the system cannot be implemented on a combined cycle plant without significant negative impact on fuel consumption and therefore efficiency. Most of the implementations outlined in the '158 patent use a recuperator to heat the air in simple cycle operation, which mitigates the fuel consumption increase issue, however, it adds significant cost and complexity. The proposed invention outlined below addresses both the cost and performance shortfalls of the invention disclosed in the '158 patent.
Also, as outlined in a related U.S. Pat. No. 5,934,063 to Nakhamkin (the “'063 patent”), there is a valve structure that “selectively permits one of the following modes of operation: there is a gas turbine normal operation mode, a mode where air is delivered from the storage system and mixed with air in the gas turbine, and then a charging mode”. The '063 patent has also been issued for more than ten (10) years and there are also no known applications of it anywhere in the world. The reason for this is again cost and performance shortfalls, similar to those related to the '158 patent. Although this system can be applied without an efficiency penalty on a simple cycle gas turbine, simple cycle gas turbines do not run very often so they typically do not pay off the capital investment in a timeframe that makes the technology attractive to power plant operators. Likewise, if this system is applied to a combined cycle gas turbine, there is a significant heat rate penalty, and again the technology does not address the market needs. The proposed invention outlined below addresses both the cost and performance issues of the '063 patent.
Gas Turbine (GT) power plants provide a significant amount of power to the grid and are used for both base load capacity and regulation on the grid. Because of fluctuating electrical load demand and fluctuations in renewable energy supply, the GT power plants are required to change load frequently. Typically, the grid operator, who is monitoring the demand, supply and frequency of the grid, sends a signal to the gas turbine fleet on a plant-by-plant basis, to supply more or less power to make the supply meet the demand and hold frequency at 50 or 60 hz. This signal is called an Active Grid Control (AGC) signal.
FIG. 9A shows the typical load on the Midwest Iso Grid. As it can be seen in FIG. 9A, there is a minimum load of about 48,000 MegaWatts (MW) required between 2 am and 4 am, and there is an increase in demand during the morning, typically between 4:00 am and 10:00 am, as people and businesses start using more power. Accordingly, an 18,000 MW increase over 6 hours, or 50 MW per minute ramp rate is needed. According to FIG. 9A, the real time ramp rate can be much higher over short periods of time. By 10 am, the grid power is increased to about 66,000 MW and holds at that level until about 5 pm, then power demand decreases due, for example, to some businesses closing. However, there is another peak demand around 8 pm coinciding with when people arrive home and use more power from home. FIG. 9B is an output versus efficiency chart of a typical gas turbine combined cycle plant. At night many plants are shut off or turned down to minimum load due to a drop in demand. A GT's efficiency is at its lowest point at minimum load and generally increases as load increases, or about 3%, which for a F-Class gas turbine is about 200 BTU/kWh. This represents $1,600 worth of wasted fuel assuming $4/MMBTU fuel cost over an 8 hour period for a single 2×1 combined cycle F-class GT power plant (gas turbine engine and steam turbine coupled to the generator). If all 18,000 MW increase described above were from plants as described above (going from minimum load to full load), it would take seventy-two 2×1 combined cycle (CC) GT plants to go from 250 MW to 500 MW and represent $115,000 worth of wasted fuel in order to meet increasing load demands. This quantity of 2×1 GT power plants represents 36,000 MW at full load, or just a little more than half of the total 66,000 MW load based on the fact that about 40% of the installed generation in the United States is GT's. For the United States grid described above, the 66,000 MW (66 GW) represents about 16%, or one sixth of the 400 GW total US electrical demand on the grid that is supplied by GTs. Therefore, approximately $700,000 in fuel cost is wasted every day across the United States or $250,000,000 ($250 M) in one year in order to support load fluctuations on the entire grid. This is a growing challenge as renewable energy sources are becoming a bigger part of the generation and their availability can fluctuate as well.
As one skilled in the art understands, as the ramp rate of the generating asset is improved, less regulation in total is required. To support this ability to support load fluctuations, some of the grid operators pay a higher rate for the same capacity if it is able to respond faster to changing demand.